Structural modification of cuminaldehyde thiosemicarbazone increases inhibition specificity toward aflatoxin biosynthesis and sclerotia development in Aspergillus flavus

Article abstract of DOI:10.1007/s00253-017-8426-y

Aspergillus flavus is an opportunistic mold that represents a serious threat for human and animal health due to its ability to synthesize and release, on food and feed commodities, different toxic secondary metabolites. Among them, aflatoxin B1 is one of the most dangerous since it is provided with a strong cancerogenic and mutagenic activity. Controlling fungal contamination on the different crops that may host A. flavus is considered a priority by sanitary authorities of an increasing number of countries due also to the fact that, owing to global temperature increase, the geographic areas that are expected to be prone to experience sudden A. flavus outbreaks are widening. Among the different pre- and post-harvest strategies that may be put forward in order to prevent fungal and/or mycotoxin contamination, fungicides are still considered a prominent weapon. We have here analyzed different structural modifications of a natural-derived compound (cuminaldehyde thiosemicarbazone) for their fungistatic and anti-aflatoxigenic activity. In particular, we have focused our attention on one of the compound that presented a prominent anti-aflatoxin specificity, and performed a set of physiological and molecular analyses, taking also advantage of yeast (Saccharomyces cerevisiae) cell as an experimental model.

Full text of DOI:10.1007/s00253-017-8426-y

Appl Microbiol Biotechnol
DOI 10.1007/s00253-017-8426-y
APPLIED GENETICS AND MOLECULAR BIOTECHNOLOGY
Structural modification of cuminaldehyde thiosemicarbazone
increases inhibition specificity toward aflatoxin biosynthesis
and sclerotia development in Aspergillus flavus
1
1
1
2
Francesca Degola & Franco Bisceglie & Marianna Pioli & Sabrina Palmano
&
3
1
1
1
Lisa Elviri & Giorgio Pelosi & Tiziana Lodi & Francesco Maria Restivo
Received: 17 February 2017 /Revised: 29 June 2017 /Accepted: 2 July 2017
#
Springer-Verlag GmbH Germany 2017
. .
Keywords Aflatoxins Aspergillus flavus
Thiosemicarbazones Proteomics Generegulation Sclerotia
Abstract Aspergillus flavus is an opportunistic mold that rep-
resents a serious threat for human and animal health due to its
ability to synthesize and release, on food and feed commodities,
different toxic secondary metabolites. Among them, aflatoxin
B1 is one of the most dangerous since it is provided with a
strong cancerogenic and mutagenic activity. Controlling fungal
contamination on the different crops that may host A. flavus is
considered a priority by sanitary authorities of an increasing
number of countries due also to the fact that, owing to global
temperature increase, the geographic areas that are expected to
be prone to experience sudden A. flavus outbreaks are widen-
ing. Among the different pre- and post-harvest strategies that
may be put forward in order to prevent fungal and/or mycotoxin
contamination, fungicides are still considered a prominent
weapon. We have here analyzed different structural modifica-
tions of a natural-derived compound (cuminaldehyde
thiosemicarbazone) for their fungistatic and anti-aflatoxigenic
activity. In particular, we have focused our attention on one of
the compound that presented a prominent anti-aflatoxin speci-
ficity, and performed a set of physiological and molecular anal-
yses, taking also advantage of yeast (Saccharomyces
cerevisiae) cell as an experimental model.
.
.
.
Introduction
Aspergillus flavus is a filamentous fungus that affects several
important crops in many different geographic areas of the
world (Amaike and Keller 2011). Corn fields, in particular,
depending on peculiar environmental condition that have
been exacerbated by global climate change (Medina et al.
2014) are prone to develop extensive fungal contamination.
However, fungal infection is not limited to the pre-harvest
phase of corn management but it may be carried over to the
post-harvest phase of the food/feed chain. The main econom-
ic issues posed by these fungal outbreaks are strictly bound to
the ability of A. flavus to activate a set of secondary metabolic
routes that lead to the accumulation of toxic compounds,
among which aflatoxin (AF) is the prominent one owing to
its cancerogenic and hepatoxigenic activity on animals and
human beings. Efforts to tackle the AF issue have been un-
dertaken by implementing different strategies (for a review,
see Abbas et al. 2009): (a) good agronomical practices to
minimize plant and fungal stress condition that are known
to trigger aflatoxin biosynthesis, (b) breeding for resistant
corn lines, (c) biocontrol by the use of different microorgan-
isms and in particular naturally occurring atoxigenic strains of
A. flavus in order to displace the threatening strains from the
ecological niche (Ehrlich et al. 2015), and (d) fungicides to
prevent the development of fungal contamination in particular
during post-harvest corn storage (Cooper and Dobson 2007).
Fungicides are an important weapon that has been largely
used to increase food availability and food safety (Cooper
and Dobson 2007) even if, as for any anti-microbial agent,
concerns have been expressed on the possible development of
Electronic supplementary material The online version of this article
(doi:10.1007/s00253-017-8426-y) contains supplementary material,
which is available to authorized users.
*
Francesco Maria Restivo
francescomaria.restivo@unipr.it
1
2
3
Dipartimento di Scienze Chimiche, della Vita e della Sostenibilità
Ambientale, Università di Parma, Parma, Italy
Istituto per la Protezione Sostenibile delle Piante, Consiglio
Nazionale delle Ricerche, Torino, Italy
Dipartimento di Scienze degli Alimenti e del Farmaco, Università di
Parma, Parma, Italy

Appl Microbiol Biotechnol
resistance among the targeted microbial population (Ma and
Michaiides 2005; Anderson 2005). For this, periodically, new
molecules must be developed in order to protect crops and
prevent sanitary risks and economic loss for the stakeholders.
Designing new molecules with anti-fungal activity is not a
simple task; specificity, beside human and animal safety, is
one of the major issues to be taken in consideration in order to
prevent ecological risks. As far as A. flavus infection of corn
is concerned, the prevalent sanitary and economic damage
derives from AF contamination. Thus, designing molecules
that may interfere with the biosynthetic ability of the fungus
to synthesize and release the relevant mycotoxin should be
pursued. Hopefully, the goal of designing a molecule provid-
ed of specificity to the relevant pest should also be addressed
by targeting a metabolic pathway specific to A. flavus and few
other species belonging to the Aspergillus genera. The enzy-
matic steps in the AF biosynthetic pathway have substantially
been elucidated (Yu 2012), and the genes encoding for the
relevant proteins were found clustered inside a about 80-kb
DNA sequence that harbors other genes not directly involved
in AF synthesis, together with two genes encoding transcrip-
tional regulators of the AF biosynthetic genes (Ehrlich et al.
modifications to the best-performing molecule
(cuminaldehyde thiosemicarbazone) with the aim to individ-
uate a structural derivative that would possibly acquire the
desired specificity: affecting AF biosynthesis while leaving
mycelium growth unaffected. The derivatives were then test-
ed for their biological activity taking also advantage of a yeast
(Saccharomyces cerevisiae) model system to get insight into
the mechanisms that regulate fungal response to oxidative
stress conditions. A proteomic approach was also adopted
to individuate the gene products that may be affected when
A. flavus cells were treated with the best-performing structur-
al derivative. Beside the expected downregulation of genes
that belong to the AF biosynthetic pattern, we found the up-
regulation or downregulation of genes that are expected to
modulate carbon flow in response to a changing metabolic
environment. The expression of secondary metabolism and
fungal differentiation regulator was also affected by the rele-
vant treatment.
Material and methods
2
005). However, despite decades of investigations spent to
Modifications of the thiosemicarbazone chemical
structure
unravel the molecular mechanism that control AF biosynthe-
sis, a comprehensive picture is still wanting in many respects.
Presently, we are facing with an impressive amount of data
and regulatory models of AF production that bring into play
the networks controlling oxidative stress response and fungal
differentiation (Reverberi et al. 2012; Roze et al. 2013;
Amare and Keller 2014; Calvo and Cary 2015). In any case,
it is generally accepted that more research is requested to
enforce the relevant hypotheses. In a recent approach to iden-
tify new synthetic molecules that may be used to control
mycotoxin contamination in food and feed commodities, we
have compared the anti-fungal and anti-mycotoxigenic activ-
ity of various thiosemicarbazones (cuminaldehyde and
cinnamaldehyde derivatives) on two fungal species affecting
cereals (Degola et al. 2015). Thiosemicarbazones are versatile
molecules that display an interesting chemistry due to the
contemporary presence of sulfur and nitrogen (Beraldo and
Gambino 2004; Pelosi 2010). These two elements provide
thiosemicarbazones with a variety of properties that, depend-
ing on the cellular environment in which they are placed,
stretch from redox activity to radical scavenging (Bisceglie
et al. 2014). In addition, they can be easily modified at will in
order to endow them with properties of specific interest.
Among the molecules that we have synthesized until now,
the most promising ones are those with hydrophobic substit-
uents. In particular, two derivatives, namely, cuminaldehyde
and cinnamaldehyde thiosemicarbazones, displayed a strong
anti-AF activity, although a dose-dependent inhibition of
growth was also observed for A. flavus (Degola et al.
All reactants used were purchased from Sigma-Aldrich (Saint
Louis, MO, USA). The H-NMR spectra were recorded on a
1
Bruker ANOVA spectrometer (Billerica, MA, USA) at
400 MHz using tetramethylsilane (TMS) as internal reference;
all the chemical shifts are reported in ppm. The Fourier-
transform infrared (FT-IR) measurements were recorded on a
Nicolet 5PC FTIR using the compounds directly on the atten-
−1
uated total reflectance (ATR) accessory in the 4000–400-cm
range. The relative intensity of reported FT-IR signals are
defined as s = strong, br = broad, m = medium, and w = weak.
Melting points were determined with a Gallenkamp instru-
ment (Weiss-Gallenkamp, Loughborough, UK). The desired
thiosemicarbazones (and the semicarbazone; see Fig. 1 for the
relevant structures) were obtained mixing equimolar amounts
of thiosemicarbazide (or semicarbazide) with the appropriate
aldehyde in ethanol. The mixtures were refluxed under stirring
for 8 h and left overnight at 0 °C. The precipitates were filtered
out, washed with cold ethanol, and dried under vacuum.
Additional details of the procedures are reported in the online
supplementary material.
Microbial strains
A. flavus Fri2+ strain (MUT 5849), which is a representative
strain of a large laboratory collection obtained sampling corn
fields of the Po Valley (Italy), was selected for this study since
it has already been used for the previous experimentation
(Degola et al. 2015) concerning thiosemicarbazone activity
2
015).In the present paper, we have applied some chemical

Appl Microbiol Biotechnol
Fig. 1 Structures of the parent
compound, cuminaldehyde (1),
and of the derivatives.
Cuminaldehyde
thiosemicarbazone (Htcum, 2), 2-
isopropylbenzaldehyde
thiosemicarbazone (mHtcum, 3),
3
-isopropylbenzaldehyde
thiosemicarbazone (oHtcum, 4),
and cuminaldehyde
semicarbazone (Hscum, 5)
1
2
3
4
5
on fungal growth and aflatoxin biosynthesis. S. cerevisiae
strains were BY4741 (MATa his3 leu2 met15 ura3)
and aflatoxin accumulation was monitored by fluorescence
emission determination with a microplate reader (TECAN
SpectraFluor Plus, Männedorf, Switzerland; settled parame-
ters λex = 360 nm, λem = 465 nm. manual gain = 83, lag
time = 0 μs, number of flashes = 3, integration time = 200 μs).
(
Brachman et al. 1998) W303-1B (MATa ade2-1 leu2-3,112
ura3-1 trp1-1 his3-11,15 can1-100) (Thomas and Rothstein
Y757C
1
989) carrying mip1
mutant allele cloned in pFL39 cen-
tromeric vector and DWM-5A (MATa ade2-1,leu2-3,112
ura3-1 trp1-1 his3-11,15 can1-100 mip1::KanR) carrying
2
,2-Diphenyl-1-picrylhydrazyl radical scavenging activity
G807R
mip1
mutant allele cloned in pFL39 centromeric vector
assay
(
Baruffini et al. 2006). Both A. flavus and S. cerevisiae strains
reported above are available from the corresponding author on
request.
Measurement of scavenging activity of thiosemicarbazones
against the 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical
was performed in accordance with Choi et al. (2002), measur-
ing the bleaching of a purple-colored methanol solution of the
stable DPPH radical. Briefly, a 4 ml solution containing the
selected compound was added to 1-ml aliquot of freshly pre-
pared 85 μM DPPH. The resulting mixture was stirred and
then incubated for 30 min at room temperature. The
thiosemicarbazone concentration was 75 μM. The decrease
in absorbance measured at 518 nm depicts the scavenging
activity of the compound against DPPH. Ascorbic acid
Evaluation of A. flavus mycelial growth and aflatoxin
accumulation in the medium
Conidium suspensions were obtained from 10-day YES agar
[
2% (w/v) yeast extract (Difco, Detroit, MI, USA), 5% (w/v)
sucrose (Sigma, St Louis, MO, USA), 2% (w/v) agar (Difco,
Detroit, MI, USA)] cultures incubated at 28 °C; conidium
concentration (quantified by optical density at 620 nm) and
viability (>90%) were determined according to Degola et al.
(30 mM) was used as positive control to determine the max-
imal decrease in DPPH absorbance (Choi et al. 2002).Values
were expressed in percentage of inhibition of DPPH absor-
bance in relation to the control values without the
thiosemicarbazone (ascorbic acid maximal inhibition was
considered 100% of inhibition).
(
2011). Conidium germination rate and post-germination hy-
phal outgrowth was assessed according to Degola et al.
2015), analyzing changes in optical density of spores suspen-
sion 46 h after inoculums in a 96-well microtiter plate
Sarstedt, Newton, NC, USA) in YES liquid medium added
(
(
with thiosemicarbazones and incubated at 28 °C. Five repli-
cates per condition were inoculated.
Anti-oxidant spot assay
The effect on aflatoxin biosynthesis was assessed by the
microplate fluorescence-based procedure already described
To examine the anti-oxidant effect of the test compounds,
(
Degola et al. 2015). Briefly, suspensions of conidia were
in vitro bioassays of S. cerevisiae (BY4741 strain) were per-
formed as described by Kim et al. (2005); 1 × 10 cells/ml
cultured in YPD at 28 °C overnight were serially diluted from
6
diluted to the appropriate concentrations and brought to the
final concentration of 5 × 10 conidia/well in standard flat-
2
bottom 96-well microplates (Sarstedt, Newton, NC, USA);
cultures were set in a final volume of 200 μl/well of coconut
milk-derived medium (CCM; Degola et al. 2011, 2012) added
with thiosemicarbazones. The plates were incubated in the
dark under stationary conditions for up to 6 days at 25 °C,
10- to 100-fold in liquid synthetic medium (SD; 6.9% w/v
YNB (Formedium , Swaffham, King’s Lynn, Norfolk, UK)
supplemented with 2% (w/v) glucose and appropriated amino
acids and bases for auxotrophy), and then, cells from each
serial dilution were spotted adjacently on SD agar medium
™

Appl Microbiol Biotechnol
supplemented with 50 μM of each compound to be tested
0.5% v/v DMSO was used as control) and 0.5 or 1 mM hy-
drogen peroxide (H O ). To examine alleviation of oxidative
10 min, the samples were centrifuged for 15 min at 13,000×g at
4 °C and the pellet was washed with cold acetone; this washing
step was repeated for three times. Pellets were dried in under a
vacuum pump, then resuspended in rehydration buffer [8 M
urea, 2% (w/v) 3-[(3-cholamidopropyl)dimethylammonio]-1-
propanesulfonate hydrate (CHAPS)] and stored at −80 °C until
further use. Total proteins content was determined according to
the Bradford protein assay (Bradford 1976), using bovine se-
rum albumin as standard.
(
2
2
stress of vanillic, gallic, and dihydrolipoic acids; glutathione
GSH); and thiosemicarbazones (all supplemented at the con-
centration of 50 μM) on cells, 0.5 or 1 mM hydrogen peroxide
H O ) was incorporated into the medium. Cell growth was
(
(
2
2
monitored at 28 °C for 7 days. Each assay was performed in
triplicate. Compounds were considered to have anti-oxidant
activity if cell growth improved compared to cohorts exposed
to H O without the test compound.
Proteomic analysis
2
2
Petite frequency assay
Analysis of differentially expressed proteins in A. flavus
mHtcum-treated cultures was conducted by two-
dimensional electrophoresis (2D PAGE), as described in detail
in the online supplementary material. Briefly, 125 μg of total
proteins was loaded on 7-cm-long, pH interval 3–10, strips,
isoelectrofocused and separated in a 12% polyacrylamide gels
(second-dimension run). Gels were stained with SYPRO®
Ruby Protein Stain (BioRad, Hercules, CA, USA); 2-D gel
image elaboration and analysis was carried out with the
PDQuest software version 8.0.1 (BioRad, Hercules, CA,
USA). Three technical replicates for each of the three biolog-
ical replicates were performed, for a total of nine gels for each
class (control and mHtcum-treated samples). Spots differen-
tially expressed were manually removed from the gels, sub-
jected to an in-gel digestion, and addressed to a liquid chro-
matography–electrospray ionization–mass spectrometry (LC-
ESI-MS/MS) and LTQ-Orbitrap analyses. Mass data were
submitted to database searching using SEQUEST search en-
gine Proteom Discoverer interface (Thermo Scientific, version
1.4; Thermo Fisher Scientific, Waltham, MA, USA).
To determine the effect of mHtcum on the yeast mitochondrial
DNA (mtDNA) mutability, W303-1B and DWM-5A strains har-
Y757
G807R
boring mip1
and mip1
alleles, respectively, were pre-
grown on solid synthetic complete medium [SC; 6.9% (w/v)
™
YNB (Formedium , Swaffham, King’s Lynn, Norfolk, UK),
0.1% (w/v) dropout mix] according to Baruffini et al. (2010)
supplemented with 2% (v/v) ethanol at 28 °C to counterselect
the petite cells that could be present in the population. After
approximately 60 h, strains were grown on liquid SC medium
supplemented with 2% (w/v) glucose and mHtcum (50 μM) or
dihydrolipoic acid (50 μM) for two 24-h growth cycles. Control
experiment in which equal amounts of DMSO (0.5% v/v) or
ethanol (0.1% v/v)—the solvents used to solubilize mHtcum
and dihydrolipoic acid, respectively—were added to untreated
cells was done in parallel. Cells were then counted, diluted, and
plated (200–250 cells/plate) on SC medium supplemented with
0.3% (w/v) glucose and 2% (v/v) ethanol. Petite frequency was
defined as the percentage of colonies showing the petite pheno-
type after a 5-day incubation at 28 °C. Each experiment was
repeated at least three times on two independent clones for each
strain. Statistical analysis of petite frequency was performed by a
two-tailed Z test.
RNA extraction and gene expression analysis (quantitative
real-time PCR)
Total RNA was extracted from 96-h-old microplate CCM cul-
®
Preparation of A. flavus total protein extracts
tures using TRIzol Kit (Sigma-Aldrich, Saint Louis, MO,
USA), according to the manufacturer’s instructions. Two-
hundred micrograms of mycelium was flash frozen in liquid
nitrogen, ground to a powder with an amalgamator (TAC 200/
S, Linea TAC s.r.l., Asti, Italy; oscillation frequency 4.200 p/m),
Mycelium (conidia were incubated for 4 days in the same cul-
ture conditions used for aflatoxin determination) was manually
detached from wells and rinsed in ultrapure distilled water.
Samples (200 mg each) were weighed into a prepared tube with
®
and overlaid with 300 μl of TRIzol reagent. The RNA quality
2
00 μl of glass microbeads, frozen in liquid nitrogen, ground
into a powder with the action of a dental amalgamator (TAC
00/S, Linea TAC s.r.l., Montegrosso d’Asti, Italy), and added
was confirmed by gel electrophoresis (1.5% agarose) and con-
centrations were measured using a BioPhotometer (Eppendorf,
Hamburg, Germany). The A260/A280 ratio was measured. Two
micrograms of total RNA sample was reverse-transcribed using
the Maxima First-Strand cDNA Synthesis Kit for qRT-PCR with
dsDNase (Thermo Fisher Scientific, Waltham, MA, USA), fol-
lowing the manufacturer instructions. The complementary DNA
(cDNA) samples were used as templates of qPCR reactions con-
ducted with ABI 7300 instrumentation ((Thermo Fisher
Scientific, Waltham, MA, USA) and SYBR® Green PCR
2
with 200 μl of lysis buffer [50 mM Tris-HCl pH 7.5, 2 M
thiourea, 7 M urea, 2% (v/v) Triton X-100, 1% dithiothreitol
(
1
DTT), 2% (w/v) soluble polyvinylpolypyrrolidone (PVPP),
mM phenylmethylsulphonylfluoride (PMSF), and 0.2% (v/
v) β-mercaptoethanol]. Samples were centrifuged twice for
0 min at 13,000×g, and then, 200 μl 45% (w/v) trichloroacetic
acid was added to the supernatant. After incubation on ice for
2

Appl Microbiol Biotechnol
Master Mix (Life Technologies, Waltham, MA, USA). Primer
sequences are reported in Supplementary Table S1. The concen-
tration of each primer pair was optimized in order to satisfy the
requirements for mRNA-level quantification according to the
Livak method (Livak and Schmittgen 2001). Primer validation
was carried out with the best amplification curve, and dissocia-
tion curves were used to confirm the good amplification of each
gene. At least three biological replicates were used to validate the
amplification specificity. Target gene expression was normalized
to the expression of tub1 gene, chosen as an internal standard.
The ΔC was calculated as C target gene − C internal standard;
Sclerotia production
5
A volume of 5 μl from a 1 × 10 spores/ml suspension of
A. flavus strain was point inoculated centrally on four replicate
CZ plates [1.5% (w/v) agar, 0.1% (w/v) di-potassium hydro-
gen phosphate, 0.05% (w/v) magnesium sulfate heptahydrate,
0.05% (w/v) potassium chloride, 0.3% (w/v) sodium nitrate,
3% (w/v) sucrose] added with mHtcum at the concentration of
50 μM. Plates were incubated up to 10 days to obtain sclerotia
production; then, the surface of colonies was scratched, and
sclerotia were manually recovered, ethanol-washed to remove
conidia, and counted. Data were recorded as number of scle-
rotia per mycelium colony.
T
T
T
−ΔΔCT
the expression level variations were then expressed as 2
(
with ΔΔC = C treatment − C control). Three biological and
T
T
T
three technical replicates per condition were performed. A nega-
tive control with no reverse transcriptase was also used to ex-
clude DNA contamination. Amplification conditions were 50 °C
for 2 min and 95 °C for 10 min; followed by 40 cycles at 95 °C
for 15 s and 60 °C for 1 min; and then, 95 °C for 15 s, 60 °C for
Statistical analysis
For statistical analyses, one-way analysis of variance
(ANOVA) was used in the Past 3.x software (Hammer et al.
2001). Results of mycelial growth, aflatoxin accumulation,
and scavenging activity assays were analyzed by Kruskal-
Wallis test, while statistical analysis of yeast petite frequency
was performed by a two-tailed Z test; differences were con-
sidered significant at p < 0.001. Data of sclerotia production
and ADH activity determination were analyzed using
Student’s t test; differences were considered to be statistically
significant when the p value was lower than 0.05.
1
min, 95 °C for 15 s, and 60 °C for 15 s. The significance of
differences between relative expression ratios of treated and con-
trol samples was determined using the Mann-Whitney test, with
a p ≤ 0.05 limit.
Spectrophotometric determination of alcohol
dehydrogenase enzyme activity
Mycelium, obtained from 4-day-old CCM 96-microplate cul-
tures, was manually detached from wells and rinsed in ultra-
pure distilled water. Samples (200 mg each) were weighed
into a prepared tube with 200 μl of glass microbeads, frozen
in liquid nitrogen, ground into a powder with the action of a
dental amalgamator (TAC 200/S, Linea TAC s.r.l.,
Montegrosso d’Asti, Italy), and homogenized in 400 μl of
extraction buffer composed of 100 mM tricine (pH 8.0),
Results
Modifications of the Htcum thiosemicarbazone chemical
structure
To shed light on the possible factors that endow our molecule,
Htcum, of its properties, we have started by modifying the two
constituting moieties: the thiosemicarbazone and the
cuminaldehyde fragments (see Fig. 1). As a first comparison,
we have synthesized the semicarbazone (Hscum), an analog of
cuminaldehyde thiosemicarbazone with the sulfur replaced by
an oxygen. This modification was envisaged in order to spec-
ulate on the role played by sulfur and oxygen, i.e., their ability
to form weak/strong hydrogen bondings and also by their redox
potentials. In general, sulfur forms weaker hydrogen bonds
than oxygen due to a lower electronegativity and is more sub-
ject to oxidation, through the formation of disulfide bridges. In
fact, thiosemicarbazones, thanks to their thione-thiol tautomer-
ism, in their thiolic form can behave as reducers and therefore
also as reactive oxygen species (ROS) generators or radical
scavengers. As a second step, we have modified the
cuminaldehyde thiosemicarbazone by synthesizing its meta-
and ortho-isomers. The choice of the ortho- (oHtcum) and
meta- (mHtcum) isopropyl derivatives of Htcum is based on
the fact that these analogs possess a hydrophobicity very
1
0 mM MgSO , 0.2% (v/v) β-mercaptoethanol, 40 mM
4
CaCl , 0.5% (w/v) polyvinylpyrrolidone, 1 mM EDTA, and
2
0
.06% (v/v) Triton X-100 in a 1.5-ml Eppendorf tube. After
two rounds of centrifugation at 10,000×g for 20 min at 4 °C,
the resulting supernatant was used for alcohol dehydrogenase
(
ADH) spectrophotometric activity detection. The reaction
mixture contained 50 mM Tris-HCl pH 8, 5 mM β-
nicotinamide adenine dinucleotide (NAD), and 3% (v/v) eth-
anol. Fifty microliters of protein extract was added to 1 ml of
reaction mixture, and the mixture was incubated at 30 °C in a
Cary 300 Scan spectrophotometer. Readings of absorbance
variation (λ = 340 nM) were performed and recorded in the
linear range. Specific enzyme activity was calculated as nmol
of NAD reduced per minute per mg of protein using a NAD
3
−1
−1
molar extinction coefficient of 6.22 × 10 mol cm . Protein
concentrations in the extracts were determined according to
the Bradford protein assay (Bradford 1976), using bovine se-
rum albumin as standard.

Appl Microbiol Biotechnol
similar to cuminaldehyde, but present different shapes. This
comparison aims at understanding if the position of the isopro-
pyl group plays a generic role as a hydrophobic fragment that
simply allows the molecule to enter the cell or if its location is
part of a more specific molecular recognition mechanism.
20, and 15% inhibition, respectively, at the highest concentra-
tion). The effect on mycelium growth of the relevant molecules
was then assayed. As shown in Fig. 2, molecules that did not
substantially affect mycotoxin biosynthesis (i.e., oHtcum,
Hscum, and cuminaldehyde) did not interfere with conidium
germination and hyphal elongation (<15% inhibition).
Interestingly, mHtcum, at difference with Htcum, did not af-
fect fungal growth significantly: <20% inhibition at the highest
concentration as compared with 88% inhibition in Htcum-
treated conidia. Experimentation described in the following sec-
tions was performed using the highest concentration (50 μM) of
the relevant compounds.
Effect of the modified thiosemicarbazones on growth
and AF accumulation by A. flavus
To compare the relative efficacy of the different modifications
of Htcum to inhibit growth and/or AF biosynthesis, conidia of
A. flavus were inoculated in the appropriate medium (YES for
germination/growth assessment and CCM for toxin accumula-
tion detection; see the BMaterial and methods^ section for de-
tails). In the concentration range tested (0–50 μM), Htcum and
mHtcum were effective in decreasing toxin accumulation (80
and 67% toxin inhibition, respectively, at the highest concen-
tration; Fig. 2). Both oHtcum, Hscum, and cuminaldehyde
only slightly reduced AF content of the culture medium (22,
Anti-oxidant activity of the modified thiosemicarbazones
Anti-oxidative stress response is considered to be a prominent
factor involved in the control of toxin accumulation in
A. flavus, and many synthetic and natural compounds that
were shown to contrast AF biosynthesis possess a ROS scav-
enging activity. The anti-oxidant activity of Htcum and of its
modified derivatives was compared by performing both
in vitro and in vivo assays.
In vitro assay
At first, the in vitro anti-oxidant activity of the various com-
pounds was tested by the DPPH assay. As shown in Fig. 3,
both mHtcum and oHtcum displayed a strong anti-oxidant
activity (i.e., comparable to the control activity; set to 100%)
and higher than that of Htcum (about 70%). This was not the
case for cuminaldehyde (no detectable anti-oxidant activity)
and for the semicarbazone, Hscum, that was provided with a
Fig. 2 Activity on mycelium growth and toxin accumulation.
Aspergillus flavus conidia were inoculated either in YES or CCM
medium for a growth determination and b aflatoxin production,
respectively (see the BMaterial and methods^ section for details). Media
were amended with 10, 25, or 50 μM of Htcum, mHtcum, oHtcum,
Hscum (see Fig. 1 for acronyms), or cuminaldehyde (cuminal.), respec-
tively, or DMSO (dimethyl sulfoxide) as control. Data are presented as
percentage of inhibition as compared to the control. Asterisk indicates the
differences that were statistically significant (p value <0.05) between
equivalent concentrations of mHtcum, oHtcum, Hscum or
cuminaldehyde, and those of Htcum. Error bars indicate the standard
deviations of results of five replications
Fig. 3 In vitro anti-oxidant assay (DPPH assay). The values of the scav-
enging activity of the compounds (see Fig. 1 for their acronyms) are
expressed in percentage of inhibition in relation to ascorbic acid
(30 mM) scavenging activity (100%). Data for the compound concentra-
tion (75 μM) are presented as mean ± SEM (n = 3). Different letters over
the bars indicate the differences that were statistically significant (p value
<0.05)

Appl Microbiol Biotechnol
low (20%) anti-oxidant activity. No consistent correlation be-
tween the anti-aflatoxigenic and the in vitro anti-oxidant ac-
tivity of the various molecules was observed. Notably, a
strong anti-oxidant activity is coupled to AF inhibition by
mHtcum, whereas the anti-oxidant activity of oHtcum is
not associated with a parallel inhibitory effect on AF biosyn-
thesis (see data reported in Fig. 2 for comparison).
shown. None of the tested molecules did affect yeast growth
when delivered to the control medium (no H O added).
2
2
Similarly, no effect was observed when the medium was
amended with a hydrogen peroxide concentration (0.5 mM)
that does not significantly affect control cultures. However,
when H O concentration in the medium was risen to 1 mM,
2
2
a dramatic inhibition of cell growth was observed, in particu-
lar, for cells that were H O -challenged during the idiophase
2
2
(no yeast colonies formed even at the highest cell concentra-
In vivo assays
tion). Amending the medium with, mHtcum or oHtcum (and
partially with Htcum) was beneficial in contrasting the oxida-
tive stress induced by hydrogen peroxide on the cells collected
during the idiophase, whereas cuminaldehyde was devoid of
efficacy. Transition of yeast cells from idiophase to
trophophase is accompanied by an increased tolerance to the
oxidative stress together with an augmented efficacy of
mHtcum, oHtcum, and Htcum in contrasting the inhibitory
activity of H O on growth. Here again, cuminaldehyde was
devoid of efficacy in contrasting the oxidative stress.
Additional evidence of the anti-oxidative efficacy of
mHtcum was obtained comparing its activity with a set of
conventional anti-oxidant compounds (Fig. 4b); mHtcum
performed even better than dihydrolipoic acid, vanillic acid,
gallic acid, and glutathione when added at equimolar concen-
tration (50 μM) to the medium.
To test the in vivo anti-oxidant efficacy of the relevant com-
pounds (Htcum, mHtcum, oHtcum, and cuminaldehyde as a
negative control), S. cerevisiae was used as a model organism
since yeast bioassay have already been shown to be predictive
of the effect of aflatoxin inhibitors on fungal cell physiology
(
Kim et al. 2005, 2008).
Spot assays Yeast cells (serial dilutions) were spotted on solid
medium containing different concentration of hydrogen per-
oxide and amended with the compounds (50 μM) to be tested
for their anti-oxidative activity. In Fig. 4a, the output of the
spot test performed on yeast cells collected during the
idiophase (during the exponential phase) and trophophase (at
the beginning of the stationary phase) of the cell culture is
2 2
Petite frequency assay To further test the hypothesis that the
antagonistic effect of mHtcum on AF may be dependent on
its anti-oxidant activity, an additional yeast assay was per-
formed, taking advantage of yeast mutants impaired in Mip1
function. In S. cerevisiae, the MIP1 gene encodes for the mi-
tochondrial DNA polymerase (POL gamma in human).
Specific mip1 point mutants, owing to a reduction of the mi-
tochondrial DNA polymerase catalytic activity, accumulate an
increased amount of mtDNA mutations that in yeast lead to
the so-called petite phenotype. Indeed, mtDNA mutants pro-
duce small colonies on media containing limiting amount of
fermentative carbon sources being unable to produce biomass
through respiration of ethanol, the end product of fermenta-
tion. It has been demonstrated that the supplementation of
anti-oxidant molecules, such as dihydrolipoic acid or MitoQ,
could rescue the petite mutability (Baruffini et al. 2006, 2012).
However, a different response of the cells to anti-oxidant treat-
ment has been observed, depending on the mutation in the
Fig. 4 Anti-oxidative stress response (yeast bioassay). Three serial
dilutions (approximately 10 , 10 , 10 cells/spot) of yeast cultures
4
3
2
(
upper and lower rows of spots corresponding to the idiophase and to
MIP1 gene. The petite frequency of cells carrying the
Y757C
the trophophase of the cell cultures, respectively) were spotted on the
solid medium containing a two hydrogen peroxide concentrations (0.5
and 1 mM) and various thosemicarbazones (50 μM of mHtcum,
oHtcum, Htcum or cuminaldehyde (cuminal.) respectively, or 0.5%
DMSO as control). In b, mHtcum (50 μM) anti-oxidant activity is com-
pared to that of various natural anti-oxidants: dihydrolipoic acid (lipoic
ac.), gallic acid (gallic ac.), and vanillic acid (vanillic ac.) or glutathione
mip1
mutation is reduced by treatment with
dihydrolipoic acid. This is not the case for cells bearing the
G807R
mip1
mutation where dihydrolipoic acid is not able to
rescue the petite mutability (Baruffini et al. 2012). Thus, the
different response to anti-oxidant molecules of the mutant
strains carrying the two mip1 mutations gives the opportunity
to challenge the hypothesis of an anti-oxidant stress-depen-
dent activity of mHtcum by measuring the petite mutability
(
GSH). Yeast culture conditions were as reported above, whereas a single
concentration (1 mM) was tested. Plates were incubated for 7 days
at 28 °C
2 2
H O

Appl Microbiol Biotechnol
Y757C
G807R
of either mip1
and mip1
mutants after treatment
with mHtcum (and dihydrolipoic acid as control). Results of
this assay are reported in Table 1. The supplementation of
mHtcum determined a strong reduction of petite colonies in
Y757C
G807R
both mip1
and mip1
mutants (52 and 38% of the
untreated control, respectively), at difference with
dihydrolipoic acid that is active only in the case of
Y757C
mip1
mutant (55% reduction of petite colonies as com-
pared to the untreated control).
Modification of A. flavus proteome by mHtcum treatment
Visualization of protein differentially induced by mHtcum
treatment was performed by 2D gel electrophoresis (2DE),
and identification by LTQ orbitrap and LC-ESI-MS/MS anal-
ysis. A total of 434 spots were detected by computer analysis.
According to the set parameters and to Student’s t test, 28
spots were significantly differentially expressed among
mHtcum amended and control cultures. Sixteen spots were
upregulated and 12 downregulated. Fourteen protein spots
Fig. 5 Gel showing the localization of the differentially expressed spots
in 50-μM-mHtcum-treated cultures. Circles show the position of the
identified spots. Protein extracts were separated on 7-cm-long IPG strips
with a pH gradient from 3 to 10, followed by 12% SDS-PAGE. The
protein spot number refers to Table 2, in which detailed information for
each protein is provided
the downregulation of OmtB in mHtcum-treated A. flavus
cells fit well with the observed repression of AF biosynthesis
by the relevant thiosemicarbazone. At a closer look, most
polypeptides that were downregulated or upregulated by
mHtcum treatment belong to carbon/energy metabolism.
(
Fig. 5) were successfully identified (Table 2). Of these 14
spots, 9 corresponded to unique proteins. Four spots sharing
similar MW but differing in their isoelectric point match the
same polypeptide, aldehyde dehydrogenase (AldA; B8N8T4),
suggesting that the relevant spots were affected by different
post-translational modifications. Similarly, two spots were
identified as ATP synthase subunit beta (B8NWC9) and two
spots as zinc-binding oxidoreductase (B8NWT2). The other
spots were identified as alcohol dehydrogenase 1 (Adh1;
P41747), transaldolase (I7ZW89), two possible malate dehy-
drogenase gene products (B8ND04 and I8TTW0, respective-
ly), and one protein as dimethyl sterigmatocystin 6-O-methyl-
transferase (Q9P900), belonging to the aflatoxin biosynthetic
pathway (OmtB/aflO gene). One spot was identified as an
uncharacterized protein of A. oryzae (I7ZL55). A BLAST
search against the UniProt KB database found a homologous
Validation studies
To validate the proteomic results, we analyzed, by reverse
transcriptase (RT) and quantitative real-time PCR (qRT-
PCR), the expression of some of the genes encoding the rele-
vant proteins. A subset of genes that are involved in the reg-
ulation of AF biosynthesis were also considered. Specifically,
using qRT-PCR, we examined the expression of aflO (OmtB
protein), aflS, aflD, veA, laeA, nsdC, nsdD, and adh1 in
A. flavus grown either in control condition or in mHtcum-
amended medium (Fig. 6). In accordance with the proteomic
analysis, aflO and adh1 were downregulated in mHtcum-
treated cultures. In this culture conditions, two more genes
belonging to the aflatoxin biosynthetic gene cluster, aflD
and aflS, were also downregulated. aflD and aflO encode
two enzymes that catalyze an early and a later step, respec-
tively, of the AF biosynthetic pathway, whereas AflS, in con-
junction with AflR, is an inherent regulator of many AF bio-
synthetic genes. The downregulation of aflO, aflD, and aflS
by mHtcum is then coherent with the reported inhibition of
AF production by mHtcum treatment. In contrast, four genes
that are known to be involved both in the regulation of AF
biosynthesis and in the control of fungal differentiation did not
respond to mHtcum treatment (veA, laeA, and nsdC) or were
slightly downregulated (nsdD). Additionally, for the alcohol
dehydrogenase protein, the spectrophotometric activity of the
enzyme was determined. Total ADH activity of A. flavus
(
100% identity) uncharacterized protein (B8NSV9) for
A. flavus. Overall, five proteins (OmtB, Adh1, the
transaldolase, and the two malate dehydrogenases) were
downregulated and four proteins (AldA, the ATP synthase
beta subunit, the zinc-binding oxidoreductase, and the
uncharacterized protein) were upregulated, in A. flavus, by
mHtcum treatment as compared to the control. Interestingly,
Table 1 Effect of mHtcum on petite induction in two mip1 yeast
mutants
Strain
mHtcum DMSO
Dihydrolipoic acid EtOH
Y757C
mip1
mip1
9.8 ± 1.3 20.5 ± 0.2 10.5 ± 0.4
5.0 ± 1.2 8.3 ± 0.1 9.2 ± 1.7
23.6 ± 3.3
9.0 ± 0.4
G807R
Petite frequency is reported as the percentage of colonies showing the
petite phenotype after a 5-day incubation at 28 °C

Appl Microbiol Biotechnol
Table 2 Differentially expressed polypeptides
No. SSP Accession Description
Score Coverage Number of Number of
Number of Number Number MW
protein
unique peptides peptides of PSMs of AAs (kDa)
1
↓
3604 Q9P900 Dimethyl sterigmatocystin
63.37 54.92
108.20 79.46
22
10
12
19
27
35
63
386
331
43.1
34.7
a
6
-O-methyltransferase
OS = Aspergillus flavus
OMTB_ASPFL)
(
2↓
6514 B8ND04 Malate dehydrogenase,
NAD-dependent
28
OS = Aspergillus flavus
(B8ND04_ASPFN)
3↓
4↓
5↓
6↑
7↑
8↑
7407 I8TTW0 Malate dehydrogenase
69.52 80.88
15.73 31.71
8.17 42.60
27
1
11
2
24
10
6
46
11
9
340
350
223
517
517
497
35.7
37.0
25.1
55.4
55.4
53.9
OS = Aspergillus oryzae
(I8TTW0_ASPO3)
6601 P41747
4403 I7ZL55
Alcohol dehydrogenase 1
OS = Aspergillus flavus
(ADH1_ASPFN)
Uncharacterized protein
2
3
OS = Aspergillus oryzae
(I7ZL55_ASPO3)
2704 B8NWC9 ATP synthase subunit β
OS = Aspergillus flavus
158.71 80.08
226.65 84.14
65.54 61.37
12
11
10
18
22
10
28
30
22
67
94
36
(B8NWC9_ASPFN)
2706 B8NWC9 ATP synthase subunit β
OS = Aspergillus flavus
(B8NWC9_ASPFN)
4706 B8N8T4 Aldehyde dehydrogenase
AldA, putative
OS = Aspergillus flavus
(B8N8T4_ASPFN)
9
↑
4712 B8N8T4 Aldehyde dehydrogenase
AldA, putative
170.35 72.43
115.61 68.41
43.32 40.85
31.69 59.13
10
10
9
23
18
4
35
30
16
16
25
13
77
63
26
21
52
49
497
497
497
323
323
324
53.9
53.9
53.9
35.7
35.7
35.5
OS = Aspergillus flavus
(B8N8T4_ASPFN)
1
0↑ 5707 B8N8T4 Aldehyde dehydrogenase
AldA, putative
OS = Aspergillus flavus
(B8N8T4_ASPFN)
1
1
1
1↑ 5712 B8N8T4 Aldehyde dehydrogenase
AldA, putative
OS = Aspergillus flavus
(B8N8T4_ASPFN)
2↑ 5608 B8NWT2 Zinc-binding oxidoreductase,
3
6
putative
OS = Aspergillus flavus
(B8NWT2_ASPFN)
3↑ 6501 B8NWT2 Zinc-binding oxidoreductase, 109.01 80.80
4
16
12
putative
OS = Aspergillus flavus
(
B8NWT2_ASPFN)
4↑ 4607 I7ZW89 Transaldolase
OS = Aspergillus oryzae
I7ZW89_ASPO3)
1
26.39 24.69
8
(
a
Variation in spot abundance between 50-μM mHtcum treatment and the corresponding control (0.5% DMSO) is indicated with a upward or downward
pointing arrow for upregulated and downregulated peptides, respectively. Fold change was considered statistically significant when it exceeds both a fold
of variation >2 and a p < 0.05
protein extracts was affected in mHtcum-amended fungal
cultures (65% inhibition as compared to control; see
Supplementary Fig. S1), consistently with adh1’s transcript
and ADH1’s polypeptide downregulation.
Modulation of sclerotia production by mHtcum
Fungal secondary metabolism is associated with development
and, in particular, AF production and sclerotia formation

Appl Microbiol Biotechnol
tested for their possible fungistatic activity or their ability
to prevent mycotoxin contamination of food and feed com-
modities (Cooper and Dobson 2007; Cantrell et al. 2012).
In particular, targeting the mycotoxin biosynthetic apparatus
rather than preventing mycelia growth may be considered
the eligible strategy when, as is the case of aflatoxins, toxin
accumulation on the commodities represents the prominent
sanitary and economical risk. However, individuation of a
molecular target, specific for the AF biosynthetic apparatus,
to be addressed by an inhibitor requires a good knowledge
of the regulatory and biosynthetic mechanisms that control
AF production by the fungus. Alternatively (or complemen-
tary to the above reported approach), a Bshotgun^ analysis
of the anti-aflatoxigenic efficacy of a set of molecules dif-
fering in their chemical structure/activity may be tried. In
the present paper, taking advantage of preliminary experi-
mentation (Degola et al. 2015), we have synthesized a set
of analogs of cuminaldehyde-thiosemicarbazone (Htcum)
molecule that displayed a strong anti-aflatoxigenic activity
albeit paralleled by a fungistatic activity. To start with, we
have examined the role of sulfur by including in the pool
of molecules to be tested the oxygen analog of the parent
molecule, cuminaldehyde semicarbazone (Hscum), but the
effects of the substitution on the proliferation and the afla-
toxin accumulation resulted to be dramatically reduced. A
second question in the spotlight was if the activity of the
compound was related to the simple hydrophobic nature of
the molecule or to a more specific structural feature. To this
purpose, we have synthesized the ortho- and meta-analogs
of cuminaldehyde, in which the isopropyl group is found in
para-position with respect to the aldehyde functional group.
The meta-isopropyl derivative (mHtcum) of Htcum
emerged among the others as the Bgood candidate^: main-
taining a remarkable anti-aflatoxigenic activity while losing
most of its fungistatic efficacy. Modulation of oxidative
stress response of A. flavus has been invoked as being
involved in the efficacy of various inhibitors of AF biosyn-
thesis (Reverberi et al. 2005; Nesci et al. 2008; Kim et al.
Fig. 6 Fold change of gene expression in response to mHtcum
treatment. A. flavus conidia were incubated and cultivated in the
identical cultural conditions (medium amended with 50 μM mHtcum)
used for aflatoxin determination and proteomic analysis (see legend of
Fig. 2). The 1.0 line represents the control (0.5% DMSO) expression
level. Asterisk indicates the differences that were statistically significant
(p value <0.05)
share various regulators such as VeA, LaeA, NsdC, and NsdD
Amare and Keller 2014; Calvo and Cary 2015). We were
(
prompted to test whether mHtcum, beside inhibiting AF bio-
synthesis, may affect sclerotia production in A. flavus. As
shown in Fig. 7, inclusion of mHtcum in the medium dras-
tically reduced the number of sclerotia as compared to un-
treated control cultures (to <10% of control culture).
Discussion
Fungicides are important tools for fungal disease control,
and several synthetic and natural compounds have been
2
008; Passone et al. 2012; Ferreira et al. 2013; Chitarrini
et al. 2014; Yan et al. 2015; Liang et al. 2015; Sun et al.
015; Jahanshiri et al. 2015; Caceres et al. 2016). We test-
2
ed this hypothesis performing both an in vitro assay and
in vivo using a yeast model system that has already been
validated by Kim (Kim et al. 2005, 2008). The results of
these analyses indicate that mHtcum is provided with an
anti-oxidant activity, comparable to that displayed by natu-
ral molecules such as dihydrolipoic acid, vanillic acid, or
glutathione. However, when a more stringent assay was
performed by the use of two yeast mutants of the same
Fig. 7 mHtcum effect on sclerotia development. A. flavus conidia were
point spotted in Czapek solid medium (amended with 50 μM mHtcum or
gene (MIP1) that are differently responsive to anti-oxidant
0
.5% DMSO as a control) and plates incubated at 28 °C for up to 10 days.
Statistically significant changes (as compared to the control mean value
10 sclerotia/colony) are indicated with asterisks (**p value <0.001).
Error bars indicate the standard deviations of three replications
Y757C
treatment (mip1
rescues the wild-type phenotype when
treated with dihydrolipoic acid or mitoQ, whereas
1
G807R
mip1
does not have benefit by a similar treatment),

Appl Microbiol Biotechnol
mHtcum was shown to affect equally both mutants. These
results give evidence that mHtcum may have an additional
target that is not directly related to its anti-oxidant activity.
To enforce the hypothesis and to obtain indications of
the nature of a possible additional target for mHtcum, not
dependent on its anti-oxidant activity, a proteomic analysis
was performed. We were able to individuate and character-
ize 16 polypeptides differently expressed in the mHtcum-
treated cultures. The results show that at least one gene
product (OmtB; aflO) belonging to the AF biosynthetic
pathway was downregulated. Subsequent gene expression
analysis confirmed the downregulation of aflO transcript
by mHtcum treatment and showed that two more genes
belonging to the AF gene cluster, aflD (which encode for
an enzyme involved in aflatoxin biosynthesis) and aflS
Fig. 8 Schematic representation of the regulatory pathways involved in
aflatoxin biosynthesis and sclerotia differentiation in A. flavus. The
speculative sketch depicts the possible targets for mHtcum activity.
Only the regulatory factors analyzed in the present paper (VeA, LaeA,
NsdC, NsdD, and AflS) are represented in the scheme. Some of the
enzymes belonging to the aflatoxin pathway or to carbon/energy metab-
olism that are upregulated or downregulated by mHtcum treatment are
also reported. The still undeciphered regulatory elements/interactions of
the pathways are indicated as a question mark inside a Bblack box^
(
which, along with aflR, display a regulatory role in myco-
toxin production), were downregulated. Altogether, these
data give evidence that mHtcum influence AF production
modulating the transcription of genes that are directly in-
volved in the biosynthetic process. However, we cannot rule
out the possibility that the target of mHtcum is upstream of
the regulatory genes located in the AF gene cluster. In this
sense, it should be noted that mHtcum treatment not only
inhibits AF production but severely impairs also sclerotia
development in A. flavus. This result may have been ex-
pected since AF biosynthesis and sclerotia production have
been shown to share several regulatory steps (Amare and
Keller 2014). We were then prompted to analyze the ex-
pression of genes encoding for transcription factors that are
supposed to display their role upstream of mycotoxin bio-
synthetic and sclerotia development pathways. veA and
laeA belong to the so called Bvelvet complex^ and are
involved in the regulation of secondary metabolism and
development in several Aspergillus species (Bayram and
Braus 2012). Indeed, deletion of either veA or laeA genes
has been reported to inhibit AF accumulation and sclerotia
development in A. flavus (Bok and Keller 2004; Duran
et al. 2007). Our results show that mHtcum treatment does
not affect significantly the expression of veA and laeA in
A. flavus, giving thus evidence that these genes are not the
target of Htcum activity. Besides veA and laeA, other tran-
scription factors such as nsdC and nsdD are involved in the
regulation AF production and morphogenesis (Cary et al.
possible involvement of nsdD in the molecular mechanism
switched on by mHtcum treatment. Unfortunately, details,
for A. flavus, of the molecular interactions of NsdD with
other regulatory factors are not available (to our knowledge)
at present. Recently, eugenol, a natural anti-oxidant, was
shown to inhibit AF biosynthesis while leaving fungal
growth mostly unaffected (Caceres et al. 2016). In this case,
the authors gave evidence that this inhibition was mainly
governed by the modification of mtfA (a putative C H zinc
2
2
finger transcription factor) and veA expression (Caceres
et al. 2016). Even if we did not analyze mtfA gene expres-
sion in the present work, based on our data on veA tran-
scription, we must conclude that a different target (from that
affected by eugenol treatment) should be involved in
mHtcum activity; however, we cannot exclude that the spe-
cific cultural conditions (media, temperature, time of incu-
bation) we have used may have led to different results from
those reported by Caceres and coworkers (Caceres et al.
2016).
We then focalized our attention on the gene products, not
belonging to the AF cluster, whose expression resulted to be
affected by mHtcum treatment. Interestingly, several pro-
teins were identified as being enzymes involved in carbon/
energy metabolism (alcohol dehydrogenase, aldehyde dehy-
drogenase, transaldolase, malate dehydrogenases, ATP syn-
thase). Moreover, both ADH and aldehyde dehydrogenase
gene expressions have been reported as being affected by
aflastatin (a potent inhibitor of AF biosynthesis) treatment
in A. flavus (Kondo et al. 2001), and these changes in the
gene transcript level were paralleled by increased ethanol
production and glucose consumption by the cells (Kondo
2
012; Gilbert et al. 2016). Here again, as was the case for
veA and laeA, nsdC transcription was not affected by
mHtcum treatment, whereas, in the same experimental con-
dition, nsdD was slightly downregulated. Altogether, these
data point to a target for mHtcum located along a different
pathway from that where veA, laeA, and nsdC are supposed
to intervene or, alternatively, downstream of these three reg-
ulators but, in any case, upstream of the step where aflS is
committed (a graphic model of the possible regulatory
interactions is presented in Fig. 8). Our data suggest a

Appl Microbiol Biotechnol
Compliance with ethical standards This article does not contain any
studies with human participants or animals performed by any of the
authors.
et al. 2001). Additionally, an interesting output of the tran-
scriptional analysis performed by Roze and coworkers (Roze
et al. 2010) on a ΔveA mutant of Aspergillus parasiticus
(
which consequently is impaired in mycotoxin biosynthesis)
Conflict of interest The authors declare that they have no competing
was the increase of expression, in the mutant as compared to
the wt strain, of a DNA sequence sharing homology to the
S. cerevisiae alcohol dehydrogenase (Adh1) gene. This may
seem contradictory when compared with our results;
mHtcum treatment, which decreased both AF and sclerotia
accumulation in A. flavus, determines the underexpression of
ADH1 (at the transcription, translation, and enzyme activity
level). However, as reported in Fig. 4, mHtcum treatment
did not affect veA gene transcription in A. flavus. Moreover,
paradoxical effects of veA on AF biosynthesis have already
been shown; as an example, AF inhibition may be achieved
either by shutting off veA expression or by treatments that
readily increase the level of veA, as is the case of the already
described effect of eugenol (Caceres et al. 2016).
Interestingly, among the other proteins affected by
mHtcum treatment, transaldolase, besides being part of a
carbon-related metabolic route, has been reported to be an
oxidative stress responsive protein in A. flavus (Zaccaria
et al. 2015). Redirection of carbon flow inside the cell,
modulating the activity of enzymes such as alcohol dehydro-
genase, aldehyde dehydrogenase, and transaldolase, may im-
pact on AF biosynthesis since the prominent substrate for
mycotoxin synthesis, the polyketide moiety, is dependent on
acetate/acetyl CoA availability in the cell (Yabe and
Nakajima 2004). Altogether, our results strengthen the hy-
pothesis that an interplay between the redox and carbon/
energy status of the fungal cell is involved in controlling
AF production and fungal differentiation. On the other hand,
AF synthesis, as a possible additional source of intracellular
ROS (Roze et al. 2015), is emerging as being part of a
general defense/signaling strategy that A. flavus put forward
to respond to a metabolic turmoil that may arise (either in
culture or field conditions) in different moments of his life
span. As a final conclusion, we would like to emphasize that
interests.
References
Abbas HK, Wilkinson J, Zablotowicz R, Accinelli C, Abel C, Bruns H,
Weaver M (2009) Ecology of Aspergillus flavus, regulation of afla-
toxin production, and management strategies to reduce aflatoxin
contamination of corn. Toxin Rev 28:142–153. doi:10.1080/
15569540903081590
Amaike S, Keller NP (2011) Aspergillus flavus. Annu Rev Phytopathol
49:107–133. doi:10.1146/annurev-phyto-072910-095221
Amare MG, Keller NP (2014) Molecular mechanisms of Aspergillus
flavus secondary metabolism and development. Fungal Genet Biol
6
6:11–18. doi:10.1016/j.fgb.2014.02.008
Anderson JB (2005) Evolution of antifungal-drug resistance: mechanisms
and pathogen fitness. Nat Rev Microbiol 3:547–556. doi:10.1038/
nrmicro1179
Baruffini E, Lodi T, Dallabona C, Puglisi A, Zeviani M, Ferrero I (2006)
Genetic and chemical rescue of the Saccharomyces cerevisiae phe-
notype induced by mitochondrial DNA polymerase mutations asso-
ciated with progressive external ophthalmoplegia in humans. Hum
Mol Genet 15:2846–2855
Baruffini E, Ferrero I, Foury F (2010) In vivo analysis of mtDNA repli-
cation defects in yeast. Methods 51:426–436
Baruffini E, Serafini F, Ferrero I, Lodi T (2012) Overexpression of DNA
polymerase zeta reduces the mitochondrial mutability caused by
pathological mutations in DNA polymerase gamma in yeast. PLoS
One 7:19–28. doi:10.1371/journal.pone.0034322
Bayram Ö, Braus GH (2012) Coordination of secondary metabolism and
development in fungi: the velvet family of regulatory proteins.
FEMS Microbiol Rev 36:1–24. doi:10.1111/j.1574-6976.2011.
0
0285.x
Beraldo H, Gambino D (2004) The wide pharmacological versatility of
semicarbazones, thiosemicarbazones and their metal complexes.
Mini Rev Med Chem 4:31–39. doi:10.2174/1389557043487484
Bisceglie F, Pinelli S, Alinovi R, Goldoni M, Mutti A, Camerini A, Piola
L, Tarasconi P, Pelosi G, Zwack PJ (2014) Cinnamaldehyde and
cuminaldehyde thiosemicarbazones and their copper(II) and
nickel(II) complexes: a study to understand their biological activity.
J Inorg Biochem 140:111–125
(
1) the molecules reported in this paper have the advantage
Bok JW, Keller NP (2004) LaeA, a regulator of secondary metabolism in
Aspergillus spp. Eukaryot Cell 3:527–535. doi:10.1128/EC.3.2.527-
of being much smaller and of facile high-yield syntheses if
compared to AF inhibitors such as blasticidin S (Yoshinari
et al. 2010, 2013) and, even more so, to polyketide analogs
such as aflastatin (Sakuda et al. 2000; Kondo et al. 2001)
and (2) mHtcum may provide an additional weapon in the
battle against mycotoxin contamination; indeed, the inhibi-
tion of sclerotia development that accompanies the anti-
aflatoxigenic activity of mHtcum may limit the survival
and dissemination of A. flavus along the food chain.
5
35.2004
Brachman CB, Davies A, Cost GJ, Caputo E, Li J, Hiater P, Boeke JD
1998) Designer deletion strains derived from Saccharomyces
(
cerevisiae S288C: a useful set of strains and plasmids for PCR-
mediated gene disruption and other applications. Yeast 14:115–132
Bradford MM (1976) A rapid and sensitive method for the quantitation of
protein utilizing the principle of protein-dye binding. Anal Biochem
7
2:248–254. doi:10.1016/0003-2697(76)90527-3
Caceres I, El Khoury R, Medina Á, Lippi Y, Naylies C, Atoui A, El
Khoury A, Oswald I, Bailly J-D, Puel O (2016) Deciphering the
anti-aflatoxinogenic properties of eugenol using a large-scale q-
PCR approach. Toxins 8:123. doi:10.3390/toxins8050123
Calvo AM, Cary JW (2015) Association of fungal secondary metabolism
and sclerotial biology. Front Microbiol 6:62. doi:10.3389/fmicb.
2015.00062
Acknowledgements Funding has been provided for this research by
Fondazione CARIPLO (Grant. No 2014-0555). We are indebted to
Matteo Manfredini and Gessica Gorbi for their help on statistical anal-
ysis, to Enrico Baruffini for providing the MIP1 yeast mutants, and to
Antonietta Cirasolo and Roberto Silva for their technical help.

Appl Microbiol Biotechnol
Cantrell CL, Dayan FE, Duke SO (2012) Natural products as sources for
new pesticides. J Nat Prod 75:1231–1242. doi:10.1021/np300024u
Cary JW, Harris-Coward PY, Ehrlich KC, Mack BM, Kale SP, Larey C,
Calvo AM (2012) NsdC and NsdD affect Aspergillus flavus mor-
phogenesis and aflatoxin production. Eukaryot Cell 11:1104–1111.
doi:10.1128/EC.00069-12
Chitarrini G, Nobili C, Pinzari F, Antonini A, De Rossi P, Del Fiore A,
Procacci S, Tolaini V, Scala V, Scarpari M, Reverberi M (2014)
Buckwheat achenes antioxidant profile modulates Aspergillus flavus
growth and aflatoxin production. Int J Food Microbiol 189:1–10.
doi:10.1016/j.ijfoodmicro.2014.07.029
Choi CW, Kim SC, Hwang SS, Choi BK, Ahn HJ, Lee MY, Park SH,
Kim SK (2002) Antioxidant activity and free radical scavenging
capacity between Korean medicinal plants and flavonoids by
assay-guided comparison. Plant Sci 163:1161–1168
Cooper J, Dobson H (2007) The benefits of pesticides to mankind and the
environment. Crop Prot 26:1337–1348
the functional genomics of antioxidant-based inhibition of aflatoxin
biosynthesis. Int J Food Microbiol 122:49–60. doi:10.1016/j.
ijfoodmicro.2007.11.058
Kondo T, Sakurada M, Okamoto S, Ono M, Tsukigi H, Suzuki A,
Nagasawa H, Sakuda S (2001) Effects of aflastatin A, an inhibitor
of aflatoxin production, on aflatoxin biosynthetic pathway and glu-
cose metabolism in Aspergillus parasiticus. J Antibiot 54:650–657
Liang D, Xing F, Selvaraj JN, Liu X, Wang L, Hua H, Zhou L, Zhao Y,
Wang Y, Liu Y (2015) Inhibitory effect of cinnamaldehyde, citral,
and eugenol on aflatoxin biosynthetic gene expression and aflatoxin
B1 biosynthesis in Aspergillus flavus. J Food Sci 80:2917–2924.
doi:10.1111/1750-3841.13144
Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression
data using real-time quantitative PCR and the 2(−ΔΔ C(T)) method.
Methods 25:402–408
Ma Z, Michailides T (2005) Advances in understanding molecular mech-
anisms of fungicide resistance and molecular detection of resistant
genotypes in phytopathogenic fungi. Crop Prot 24:853–863
Medina A, Rodriguez A, Magan N (2014) Effect of climate change on
Aspergillus flavus and aflatoxin B1 production. Front Microbiol 5:
Degola F, Berni E, Restivo FM (2011) Laboratory tests for assessing
efficacy of atoxigenic Aspergillus flavus strains as biocontrol agents.
Int J Food Microbiol 146:235–243. doi:10.1016/j.ijfoodmicro.2011.
0
2.020
1–7. doi:10.3389/fmicb.2014.00348
Degola F, Dall’Asta C, Restivo FM (2012) Development of a simple and
high-throughput method for detecting aflatoxins production in cul-
ture media. Lett Appl Microbiol 55:82–89. doi:10.1111/j.1472-
Nesci A, Ferrari L, Etcheverry M (2008) Effect of synthetic antioxidants
on stored maize grain mycoflora in situ. J Sci Food Agric 88:797–
8
04
7
65X.2012.03264.x
Passone MA, Rosso LC, Etcheverry M (2012) Influence of sub-lethal
antioxidant doses, water potential and temperature on growth, scle-
rotia, aflatoxins and aflD (=nor-1) expression by Aspergillus flavus
RCP08108. Microbiol Res 167:470–477. doi:10.1016/j.micres.
Degola F, Morcia C, Bisceglie F, Mussi F, Tumino G, Ghizzoni R, Pelosi
G, Terzi V, Buschini A, Restivo FM, Lodi T (2015) In vitro evalu-
ation of the activity of thiosemicarbazone derivatives against
mycotoxigenic fungi affecting cereals. Int J Food Microbiol 200:
2
011.11.004
1
04–111. doi:10.1016/j.ijfoodmicro.2015.02.009
Pelosi G (2010) Thiosemicarbazone metal complexes: from structure to
activity. Open Crystallog J 3:16–28
Reverberi M, Fabbri A, Zjalic S, Ricelli A, Punelli F, Fanelli C (2005)
Antioxidant enzymes stimulation in Aspergillus parasiticus by
Lentinula edodes inhibits aflatoxin production. Appl Microbiol
Biotechnol 69:207–215. doi:10.1007/s00253-005-1979-1
Duran RM, Cary JW, Calvo AM (2007) Production of cyclopiazonic
acid, aflatrem, and aflatoxin by Aspergillus flavus is regulated by
veA, a gene necessary for sclerotial formation. Appl Microbiol
Biotechnol 73:1158–1168. doi:10.1007/s00253-006-0581-5
Ehrlich KC, Yu J, Cotty PJ (2005) Aflatoxin biosynthesis gene clusters
and flanking regions. J Appl Microbiol 99:518–527. doi:10.1111/j.
Reverberi M, Punelli M, Smith CA, Zjalic S, Scarpari M, Scala V,
Cardinali G, Aspite N, Pinzari F, Payne G, Fabbri A, Fanelli C
1
365-2672.2005.02637.x
Ehrlich KC, Moore GG, Mellon JE, Bhatnagar D (2015) Challenges
facing the biological control strategy for eliminating aflatoxin con-
tamination. World Mycotoxin J 8:225–233. doi:10.3920/WMJ2014.
(
2012) How peroxisomes affect aflatoxin biosynthesis in
Aspergillus flavus. PLoS One 7:1–15. doi:10.1371/journal.pone.
048097
0
1
696
Roze LV, Chanda A, Laivenieks M, Beaudry RM, Artymovich K,
Koptina AV, Awad DW, Valeeva D, Jones AD, Linz JE (2010)
Volatile profiling reveals intracellular metabolic changes in
Aspergillus parasiticus: veA regulates branched chain amino acid
and ethanol metabolism. BMC Biochem 11:33. doi:10.1186/1471-
Ferreira FD, Kemmelmeier C, Arrotéia CC, Da Costa CL, Mallmann CA,
Janeiro V, Ferreira FMD, Mossini SAG, Silva EL, Machinski M
(
2013) Inhibitory effect of the essential oil of Curcuma longa L.
and curcumin on aflatoxin production by Aspergillus flavus Link.
Food Chem 136:789–793. doi:10.1016/j.foodchem.2012.08.003
Gilbert MK, Mack BM, Wei Q, Bland JM, Bhatnagar D, Cary JW (2016)
RNA sequencing of an nsdC mutant reveals global regulation of
secondary metabolic gene clusters in Aspergillus flavus. Microbiol
Res 182:150–161. doi:10.1016/j.micres.2015.08.007
2
091-11-33
Roze LV, Hong S-Y, Linz JE (2013) Aflatoxin biosynthesis: current fron-
tiers. Annu Rev Food Sci Technol 4:293–311. doi:10.1146/annurev-
food-083012-123702
Roze LV, Laivenieks M, Hong SY, Wee J, Wong SS, Vanos B, Awad D,
Ehrlich KC, Linz JE (2015) Aflatoxin biosynthesis is a novel source
of reactive oxygen species—a potential redox signal to initiate re-
sistance to oxidative stress? Toxins 7:1411–1430. doi:10.3390/
toxins7051411
Hammer Ø, Harper DAT, Ryan PD (2001) PAST: Paleontological
Statistics Software Package for education and data analysis.
Palaeontol Electron 4(1):1–9
Jahanshiri Z, Shams-Ghahfarokhi M, Allameh A, Razzaghi-Abyaneh M
(2015) Inhibitory effect of eugenol on aflatoxin B1 production in
Aspergillus parasiticus by downregulating the expression of major
genes in the toxin biosynthetic pathway. World J Microbiol
Biotechnol 31:1071–1078. doi:10.1007/s11274-015-1857-7
Sakuda S, Ono M, Ikeda H (2000) Blasticidin A as an inhibitor of afla-
toxin production by Aspergillus parasiticus. J Antibiot (Tokyo) 53:
1265–1271
Kim JH, Campbell BC, Yu J, Mahoney N, Chan KL, Molyneux RJ,
Bhatnagar D, Cleveland TE (2005) Examination of fungal stress
response genes using Saccharomyces cerevisiae as a model system:
targeting genes affecting aflatoxin biosynthesis by Aspergillus flavus
Link. Appl Microbiol Biotechnol 67:807–815. doi:10.1007/s00253-
Sun Q, Wang L, Lu Z, Liu Y (2015) In vitro anti-aflatoxigenic effect and
mode of action of cinnamaldehyde against aflatoxin B1. Int
Biodeterior Biodegrad 104:419–425. doi:10.1016/j.ibiod.2015.07.
009
Thomas B, Rothstein R (1989) The genetic control of direct-repeat re-
combination in Saccharomyces: the effect of rad52 and rad1 on
mitotic recombination at GAL10, a transcriptionally regulated gene.
Genetics 123:725–738
0
04-1821-1
Kim JH, Yu J, Mahoney N, Chan KL, Molyneux RJ, Varga J, Bhatnagar
D, Cleveland TE, Nierman WC, Campbell BC (2008) Elucidation of

Appl Microbiol Biotechnol
Yabe K, Nakajima H (2004) Enzyme reactions and genes in aflatoxin
Yoshinari T, Sakuda S, Watanabe M, Kamata Y, Ohnishi T, Sugita-
Konishi Y (2013) New metabolic pathway for converting blasticidin
S in Aspergillus flavus and inhibitory activity of aflatoxin production
by blasticidin S metabolites. J Agric Food Chem 61:7925–7931
Yu J (2012) Current understanding on aflatoxin biosynthesis and future
perspective in reducing aflatoxin contamination. Toxins 4:1024–
1057. doi:10.3390/toxins4111024
biosynthesis. Appl Microbiol Biotechnol 64:745–755. doi:10.
1
007/s00253-004-1566-x
Yan S, Liang Y, Zhang J, Chen Z, Liu C-M (2015) Autoxidated linolenic
acid inhibits aflatoxin biosynthesis in Aspergillus flavus via oxylipin
species. Fungal Genet Biol 81:829–837. doi:10.1016/j.fgb.2014.11.
0
05
Yoshinari T, Noda Y, Yoda K, Sezaki H, Nagasawa H, Sakuda S (2010)
Inhibitory activity of blasticidin A, a strong aflatoxin production
inhibitor, on protein synthesis of yeast: selective inhibition of afla-
toxin production by protein synthesis inhibitors. J Antibiot 63:309–
Zaccaria M, Ludovici M, Sanzani SM, Ippolito A, Cigliano RA,
Sanseverino W, Scarpari M, Scala V, Fanelli C, Reverberi M
(2015) Menadione-induced oxidative stress re-shapes the oxylipin
profile of Aspergillus flavus and its lifestyle. Toxins 7:4315–4329.
doi:10.3390/toxins7104315
3
14. doi:10.1038/ja.2010.36